WO2014094814A1

WO2014094814A1 - Method of configuring an optical communication network configured for co-propagating raman amplification and an optical communication network element

Info

Publication number: WO2014094814A1
Application number: PCT/EP2012/075738
Authority: WO
Inventors: Jeroen NIJHOF; Roberto Magri
Original assignee: Telefonaktiebolaget L M Ericsson (Publ)
Priority date: 2012-12-17
Filing date: 2012-12-17
Publication date: 2014-06-26
Also published as: EP2932629A1; US20150341117A1; US9479257B2

Abstract

A method 10 of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification, the method comprising: a. selecting an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier 12; b. selecting an optical signal power, P_ch, for at least one optical communication signal to be transmitted across the optical link, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain 14; c. generating a pump signal power control signal configured to cause an optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power 16; and d. generating a signal power control signal configured to cause the optical communication signal to be delivered into the optical link at the selected optical signal power 18.

Description

METHOD OF CONFIGURING AN OPTICAL COMMUNICATION NETWORK CONFIGURED FOR CO-PROPAGATING RAMAN AMPLIFICATION AND AN

OPTICAL COMMUNICATION NETWORK ELEMENT Technical Field

The invention relates to a method of configuring an optical communication network comprising an optical link comprising a Raman amplifier configured for co-propagating Raman amplification. The invention further relates to an optical communication network element and to an optical communication network.

Background

Long spans in optical communication networks (e.g. submarine links) require Raman amplification due to its better noise characteristics. The most demanding connections, such as those to distant islands, long festoons, and hut skipping spans, require, in addition to the usual counter-propagating Raman pumping scheme, the use of the co- propagating (also referred to a forward pumping) Raman scheme. Higher order Raman pumping schemes can also be used for further performance enhancement.

The co-propagating Raman amplification scheme is the most critical since the signal power is at its highest level in the optical link and gets amplified by the Raman pump, so nonlinear effects can degrade performance if an inappropriate channel signal power is set. The co-propagating Raman amplification scheme is also affected by the 'gain saturation' phenomenon where the gain of the Raman amplifier can be reduced due to pump depletion induced by high signal power amplification; the higher the signal power, the higher the gain reduction due to pump depletion. As a result, finding the optimal channel signal power to launch an optical signal at in an optical communication network configured for co-propagating Raman amplification is a difficult task and usually needs to be done by expert personnel in the field, with many try and see steps. In addition, when the channel count changes, for example due to network upgrades, the channel launch power needs to be re-optimized because gain saturation couples the gain of the Raman amplifier with the channel count. Existing Raman amplification networks are operated to keep the gain of the Raman amplifier constant by adjusting the Raman pump power when the channel count changes. An example of this approach is reported in P.Kim and N. Park "Semi- Analytic gain control algorithm for the fiber Raman amplifier under dynamic channel reconfiguration", Optical Fiber Communication Conference, 6-11 March 2005, OTuN5.

Summary

It is an object to provide an improved method of configuring an optical

communication network comprising an optical link comprising a Raman amplifier configured for co-propagating Raman amplification. It is a further object to provide an improved optical communication network element. It is a further object to provide an improved optical communication network.

A first aspect of the invention provides a method of configuring an optical communication network. The optical communication network comprises an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification. The method comprises step a. of selecting an optical pump signal power, Pp, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold, G_max, above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier. The method comprises step b. of selecting an optical signal power, P_c , for at least one optical communication signal to be transmitted across the optical link. The optical signal power is selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain. The method comprises step c. of generating a pump signal power control signal configured to cause an optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power. The method comprises step d. of generating a signal power control signal configured to cause the optical communication signal to be delivered into the optical link at the selected optical signal power.

The method may enable the optical communication signal power to be optimized in an optical communication network where co-propagating Raman amplification is used. This may result in an improvement in the Raman link budget, enabling longer links to be used. The method may enable an optimal combination of Raman gain and optical communication signal power to be selected while avoiding increased nonlinear impairment due to distributed Raman amplification. The method may enable the optical signal power to be optimised accounting for both pump depletion and nonlinear impairments. The method may enable use of the available optical pump signal power to be maximised.

In an embodiment, step a. comprises selecting an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is no longer negligible in the Raman amplifier. The method may enable the optical communication signal power to be optimised without incurring significant levels of double Rayleigh scattering and relative intensity noise, RIN, transfer.

In an embodiment, in step a. the gain threshold is a saturated gain above which the optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is less than 0.5dB.

In an embodiment, in step a. the gain threshold is a saturated gain above which the optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is less than O. ldB.

In an embodiment, the gain threshold is a pre-selected value which is set by a manufacturer of an optical pump signal source configured to generate the optical pump signal.

In an embodiment, the gain threshold is a pre-selected value which is determined by characterisation measurements of an optical pump signal source configured to generate the optical pump signal.

In an embodiment, the gain threshold is determined in dependence on a relative intensity noise, RIN, specification of an optical pump source of the Raman amplifier. The gain threshold may be determined using equations 7 and 14 disclosed in B.Bristiel et al "New model of noise figure and RIN transfer in fiber Raman amplifiers", Photonics Technology Letters, vol. 18, no. 8, 15 April 2006, pages 980-982. The method may therefore be used both in an optical communication network having a Raman pump signal configured for negligible RIN transfer and one in which poorer quality optical pump signals having high RIN are used. In an embodiment, the gain threshold is a maximum acceptable nonlinear impairment constrained gain. In an embodiment, the gain threshold is in the range 26dB to 28dB.

In an embodiment, step b. comprises step i. of obtaining an optimal optical signal power, P_ni, for the optical communication signal at which the Q-factor of the optical communication signal is maximised in the absence of Raman amplification. Step b.

comprises step ii. of calculating a nonlinear enhancement factor, NEF. The nonlinear enhancement factor is an optical signal power reduction required to the optimal optical signal power for said maximised saturated gain to avoid an increase in a nonlinear degradation of the optical signal when Raman amplification is applied to the optical communication signal. Step b. comprises step iii. of calculating an optimal value for the optical signal power, P_c , as P_ch = P_nl - NEF . The method may enable an optimal value to be selected for the optical signal power which avoids an increase in nonlinear impairments due to distributed Raman amplification.

In an embodiment, step b. is repeated until the optical signal power converges to a stable value. This may enable the optical signal power to be selected automatically, thereby simplifying commissioning of optical communication network links having Raman amplification and reducing the need to send an expert engineer into the field to commission a link.

In an embodiment, step a. comprises checking whether G > G_max + Δ, where Δ is a hysteresis parameter. Δ is greater than a network tolerance of the communication network. Including the hysteresis parameter may enable a stable value of the optical signal power to be selected. The hysteresis parameter is larger than the network tolerance to avoid oscillations in the optical signal power but is small enough so that G is not selected to have a value which would degrade the performance, i.e. reduce the Q-factor, of the

communication network.

In an embodiment, step a. further comprises checking whether G < G_max - Δ and P_p is less than a maximum available optical pump signal power.

In an embodiment, step a. comprises initially selecting a minimum value for the optical signal power and selecting a maximum operating signal power of an optical pump source of the Raman amplifier as an initial value for the optical pump signal power. This may enable the optical signal power to be ramped up automatically from a minimum value to its optimal value.

In an embodiment, the optimal optical signal power in the absence of Raman amplification, P_ni, is obtained by determining a number of optical communication signals, N_ch, to be transmitted across the optical link and retrieving a pre-stored value of P_ni in dependence on at least one of a currently set optical signal power, Pch, and the number of optical communication signals, a characteristic of the optical fibre and a modulation format of the optical communication signals.. The pre-stored value of P_ni may therefore be retrieved according to one or more of the type of optical fibre forming the Raman amplifier, the number of optical communication signals and the modulation format of the signals.

In an embodiment, step a. comprises selecting an initial value for the optical signal power. Step b. further comprises obtaining a value for the saturated gain, G, of the Raman amplifier at the selected optical pump signal power and the selected optical signal power. Step b. comprises calculating a nonlinear enhancement factor, NEF, for the selected optical signal power, and calculating an optimal optical signal power in the absence of Raman amplification, P_ni, as P_nl = P_ch + NEF . This may enable the method to start with a manual setting of the optical signal power, by an engineer in the field, with the method then being used to optimise the optical signal power value.

In an embodiment, the saturated gain is obtained by measurement. The nonlinear enhancement factor is calculated as NEF = M * G , where M is a scaling factor in the range 0.7-0.8.

In an embodiment, steps a. and b. are repeated until a stable value for the optical signal power is obtained. The method may therefore start with a measured value of the saturated gain and iterate until the optimal optical signal power is arrived at.

In an embodiment, Mis approximately 0.8.

In an embodiment, Mis determined by calibration of the Raman amplifier either by measurement or by simulation modelling. may thereby be determined for a specific pumping scheme of the Raman amplifier.

G₀ (z)(P_p + S) In an embodiment, the saturated gain, G(z), is calculated as G(z)

P_p + G₀ (z)S where z is a distance along the optical link, S is the normalised total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier.

In an embodiment, the unsaturated gain is given by G₀ (z) = Qxp((P_p + S)(l - e^~az )) , where a is the loss coefficient of the optical link at a wavelength of the optical

communication signal. The nonlinear enhancement factor is calculated as the ratio of the effective lengths of the Raman amplifier with and without Raman amplification. The

/ 8 P (°)

normalised optical pump power is given by P =— where f_s is the optical

fp^ap^Aeff

communication signal frequency,^, is the optical pump signal frequency, g_¾ is the Raman gain of the fibre, P_p(0) is the normalised optical pump signal power at z=0, a_p is the loss coefficient of an optical fibre of the optical link at the pump signal frequency, and A_e/f is the effective area of the fibre. The normalised total optical signal power is given by

8RP, <P) where P_s(0) is the optical signal power at z=0, and a_s is the loss coefficient aA_ff

of the fibre at the optical communication signal frequency.

In an embodiment, the nonlinear enhancement factor is calculated as

In an embodiment, in step b. the optical signal power is for a plurality of optical communication signals. The method may be used for a wavelength division multiplexed optical communication network.

In an embodiment, steps a. to d. are implemented during an initial configuration of the optical communication network.

In an embodiment, steps a. to d. are implemented following a change in the number of optical communication signals. The method may therefore be run whenever the number of optical communication signals changes, to re-optimise the optical signal power.

A second aspect of the invention provides an optical communication network element comprising an optical pump signal source, optical communication signal power control apparatus, an output and a controller. The output is configured to output at least one optical communication signal and a co -propagating optical pump signal to be delivered to an optical link comprising an optical fibre configured as a Raman amplifier. The controller is configured to: a. select an optical pump signal power, P_p, such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier; b. select an optical signal power, P_c , for the at least one optical

communication signal, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain;

c. generate a pump power control signal configured to cause the optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power; and

d. generate a signal power control signal configured to cause the optical

communication signal power control apparatus to set the optical signal power of the at least one optical communication signal to the selected optical signal power.

The optical communication network element may enable the optical

communication signal power to be optimized for launch into an optical communication network where co-propagating Raman amplification is used. This may result in an improvement in the Raman link budget, enabling longer links to be used. The optical communication network element may enable an optimal combination of optical pump signal power, and thus Raman gain in the optical communication network, and optical communication signal power to be selected while avoiding increased nonlinear impairment in the optical communication network due to distributed Raman amplification. The optical communication network element may enable the optical signal power to be optimised accounting for both pump depletion and nonlinear impairments in the optical

communication network. The optical communication network element may enable use of the optical pump signal power available from the optical pump signal source to be maximised. In an embodiment, the controller is configured in step a. to select an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is no longer negligible in the Raman amplifier. This may enable the optical communication signal power to be optimised without incurring significant levels of double Rayleigh scattering and relative intensity noise, RIN, transfer in the optical communication network with which it is to be used.

In an embodiment, in step a. the gain threshold is a saturated gain above which the optical signal to noise ratio penalty caused by double Rayleigh scattering and relative 5 intensity noise transfer is less than 0.5dB.

In an embodiment, the gain threshold is a pre-selected value which is set by a 10 manufacturer of the optical pump signal source.

In an embodiment, the gain threshold is a pre-selected value which is determined by characterisation measurements of the optical pump signal source.

In an embodiment, the gain threshold is a maximum acceptable nonlinear impairment constrained gain. In an embodiment, the gain threshold is in the range 26dB to 15 28dB.

In an embodiment, the controller is configured to, in step b.:

i. obtain an optimal optical signal power, P_ni, for the optical

communication signal at which the Q-factor of the optical communication signal is maximised in the absence of Raman

20 amplification;

ii. calculate a nonlinear enhancement factor, NEF, being an optical signal power reduction required to the optimal optical signal power for said maximised saturated gain to avoid an increase in a nonlinear degradation of the optical signal when Raman amplification is

25 applied to the optical communication signal; and

iii. calculate an optimal value for the optical signal power, P_c , as

P_ck = P_nl - NEF

This may enable an optimal value to be selected for the optical signal power which avoids an increase in nonlinear impairments due to distributed Raman amplification in the 30 optical communication network into which the communication network element delivers the optical communication signal and the optical pump signal.

In an embodiment, the controller is configured to repeat step b. until the optical signal power converges to a stable value. This may enable the optical signal power to be selected automatically, thereby simplifying commissioning of optical communication network links having Raman amplification and reducing the need to send an expert engineer into the field to commission a link.

In an embodiment, the controller is configured in step a. to check whether G > G_max + Δ, where Δ is a hysteresis parameter. Δ is greater than a network tolerance of the communication network. Including the hysteresis parameter may enable a stable value of the optical signal power to be selected. The hysteresis parameter is larger than the network tolerance to avoid oscillations in the optical signal power but is small enough so that G is not selected to have a value which would degrade the performance, i.e. reduce the Q- factor, of the communication network.

In an embodiment, the controller is further configured in step a. to check whether G < G_max - Δ and to check whether P_p is less than a maximum available optical pump signal power.

In an embodiment, the controller is configured in step a. to initially select a minimum value for the optical signal power and selecting a maximum operating signal power of an optical pump source of the Raman amplifier as an initial value for the optical pump signal power. This may enable the optical signal power to be ramped up

automatically from a minimum value to its optimal value.

In an embodiment, the controller is configured in step b.i. to determine a number of optical communication signals, N_ch, to be delivered to the optical link. The controller is configured in step b.i. to obtain the optimal optical signal power, P_ni, by retrieving a pre- stored value of P_ni in dependence on at least one of a currently set optical signal power, Pch, and the number of optical communication signals, a characteristic of the optical fibre and a modulation format of the optical communication signals.. The pre-stored value of P_ni may therefore be retrieved by the controller according to one or more of the type of optical fibre forming the Raman amplifier, the number of optical communication signals and the modulation format of the signals.

In an embodiment, the controller is configured in step b. to obtain a value for the saturated gain, G, of the Raman amplifier at the selected optical pump signal power and the selected optical signal power. The controller is configured in step b. to calculate a nonlinear enhancement factor, NEF, for the selected optical signal power and to calculate an optimal optical signal power in the absence of Raman amplification, P_ni, as P_nl = P_ch + NEF . This may enable the communication network element to receive an initial manually set optical signal power, set by an engineer in the field, and then to operate to optimise the optical signal power value.

In an embodiment, the communication network element comprises gain

measurement apparatus configured to measure the saturated gain of the Raman amplifier. The controller is configured to receive the measured saturated gain and to calculate the nonlinear enhancement factor as NEF = M * G , where M is a scaling factor in the range 0.7-0.8.

In an embodiment, the controller is configured to repeat steps a. and b. until a stable value for the optical signal power is obtained. The network element may therefore start with a measured value of the saturated gain and the controller operate iteratively until the optimal optical signal power is arrived at.

In an embodiment, Mis approximately 0.8.

In an embodiment, the controller is configured to calculate saturated gain, G(z), as G₀ (z)(P_p + S)

G(z) = , where z is a distance along the optical link, S is the normalised

P_p + G₀ (z)S '

total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier.

In an embodiment, the unsaturated gain of the Raman amplifier is given by

G₀ (z) = exp((i^, + S)(l— e ^az )) , where a is the loss coefficient of the optical link at a wavelength of the optical communication signal. The controller is configured to calculate the nonlinear enhancement factor as the ratio of the effective lengths of the Raman amplifier with and without Raman amplification. The normalised optical pump signal

/ 8 P (°)

power is given by P = where f_s is the optical communication signal

frequency,^, is the optical pump signal frequency, g¾ is the Raman gain of the fibre, P_p(0) is the normalised optical pump power at z=0, a_p is the loss coefficient of the fibre at the pump signal frequency, and A_e/f is the effective area of the fibre. The normalised total optical signal power is given by S = 8 ^R P ^s (0) where P_s(0) is the optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency. In an embodiment, the ntroller is configured to calculate the nonlinear enhancement factor as NEF = .

In an embodiment, the controller is configured in step b. to select the optical signal power for a plurality of optical communication signals. The communication network element may be used in a wavelength division multiplexed optical communication network.

In an embodiment, the controller is configured to implement steps a. to d. during an initial configuration of the optical communication network.

In an embodiment, the controller is configured to implement steps a. to d. in response to a change in the number of optical communication signals. The controller may therefore re-optimise the optical signal power whenever the number of optical

communication signals changes.

In an embodiment, the optical pump signal source comprises a plurality of optical pump signal sub-sources each configured to generate an optical pump sub-signal and an optical signal combiner configured to combine the optical pump sub-signals to form the optical pump signal.

In an embodiment, the optical pump signal source is configured to generate an optical pump signal for single stage Raman pumping.

In an embodiment, the optical pump signal source is configured to generate an optical pump signal for cascaded Raman pumping.

In an embodiment, the optical communication signal power control apparatus comprises one of an optical transmitter and an optical amplifier.

A third aspect of the invention provides an optical communication network comprising an optical link and an optical communication network element. The optical link comprises an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification. The optical communication network element comprises an optical pump signal source, optical communication signal power control apparatus, an output and a controller. The output is configured to output at least one optical communication signal and a co-propagating optical pump signal to be delivered to the optical link. The controller is configured to:

a. select an optical pump signal power, P_p, such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier; b. select an optical signal power, P_c , for the at least one optical

d. generate a signal power control signal configured to cause the optical

The optical communication network may be configured for optimal optical communication signal power in the presence of co-propagating Raman amplification. This may result in an improvement in the Raman link budget, enabling longer links to be used in the network. An optimal combination of optical pump signal power, and thus Raman gain in the optical communication network, and optical communication signal power may be selected for the network while avoiding increased nonlinear impairment due to distributed Raman amplification. The optical communication network may be configured for optimal optical communication signal power accounting for both pump depletion and nonlinear impairments. Use of the optical pump signal power available from the optical pump signal source may be maximised in the network.

In an embodiment, the controller is configured in step a. to select an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is no longer negligible in the Raman amplifier. This may enable the optical communication signal power to be optimised without incurring significant levels of double Rayleigh scattering and relative intensity noise, RIN, transfer in the optical communication network. In an embodiment, in step a. the gain threshold is a saturated gain above which the optical signal to noise ratio penalty caused by double Rayleigh scattering and relative intensity noise transfer is less than 0.5dB.

In an embodiment, the gain threshold is a pre-selected value which is set by a manufacturer of the optical pump signal source.

In an embodiment, the gain threshold is a maximum acceptable nonlinear impairment constrained gain. In an embodiment, the gain threshold is in the range 26dB to 28dB.

In an embodiment, the controller is configured to, in step b.:

i. obtain an optimal optical signal power, P_ni, for the optical

communication signal at which the Q-factor of the optical communication signal is maximised in the absence of Raman amplification;

ii. calculate a nonlinear enhancement factor, NEF, being an optical signal power reduction required to the optimal optical signal power for said maximised saturated gain to avoid an increase in a nonlinear degradation of the optical signal when Raman amplification is applied to the optical communication signal; and iii. calculate an optimal value for the optical signal power, P_c , as

P_ch = P_nl - NEF

This may enable an optimal value to be selected for the optical signal power which avoids an increase in nonlinear impairments due to distributed Raman amplification in the optical communication network.

In an embodiment, the controller is configured to repeat step b. until the optical signal power converges to a stable value. This may enable the optical signal power to be selected automatically, thereby simplifying commissioning of links in the optical communication network having Raman amplification and reducing the need to send an expert engineer into the field to commission a link. In an embodiment, the controller is configured in step a. to check whether G > G_max + Δ, where Δ is a hysteresis parameter. Δ is greater than a network tolerance of the communication network. Including the hysteresis parameter may enable a stable value of the optical signal power to be selected. The hysteresis parameter is larger than the network tolerance to avoid oscillations in the optical signal power but is small enough so that G is not selected to have a value which would degrade the performance, i.e. reduce the Q- factor, of the communication network.

automatically from a minimum value to its optimal value.

In an embodiment, the controller is configured in step b. to obtain a value for the saturated gain, G, of the Raman amplifier at the selected optical pump signal power and the selected optical signal power. The controller is configured in step b. to calculate a nonlinear enhancement factor, NEF, for the selected optical signal power and to calculate an optimal optical signal power in the absence of Raman amplification, P_ni, as

P„i = P_ch + NEF . This may enable the communication network element to receive an initial manually set optical signal power, set by an engineer in the field, and then to operate to optimise the optical signal power value for the network. In an embodiment, the communication network element comprises gain

In an embodiment, Mis approximately 0.8.

G(z) = , where z is a distance along the optical link, S is the normalised

P_p + G₀ {z)S

total optical signal power and Go(z) is the unsaturated gain of the Raman amplifier.

In an embodiment, the unsaturated gain of the Raman amplifier is given by

G₀ (z) = exp((i^, + S)(l— e ^az )) , where a is the loss coefficient of the optical link at a wavelength of the optical communication signal. The controller is configured to calculate the nonlinear enhancement factor as the ratio of the effective lengths of the Raman amplifier with and without Raman amplification.

/ 8 P (°)

The normalised optical pump signal power is given by P =— where f_s is fp^ap^Aeff

the optical communication signal frequency,^, is the optical pump signal frequency, g_¾ is the Raman gain of the fibre, P_p(0) is the normalised optical pump signal power at z=0, a_p is the loss coefficient of the fibre at the pump signal frequency, and A_e/f is the effective area of the fibre. The total optical signal power is given by S = P ^s (0) where P_s(0) is the

optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency.

In an embodiment, the controller is configured to calculate the nonlinear enhancement factor as NEF

In an embodiment, the controller is configured in step b. to select the optical signal power for a plurality of optical communication signals. The optical communication network may be used for wavelength division multiplexed transmission.

communication signals transmitted across the network link changes.

A fourth aspect of the invention provides a data carrier having computer readable instructions embodied therein. The said computer readable instructions are for providing access to resources available on a processor. The computer readable instructions comprise instructions to cause the processor to perform any of the above steps of the method of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification.

In an embodiment, the data carrier is a non-transitory data carrier.

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings. Brief description of the drawings

Figure 1 shows the steps of a method according to a first embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 2 shows the steps of a method according to a second embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 3 shows the steps of a method according to a third embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 4 shows the steps of a method according to a fourth embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 5 shows the steps of a method according to a fifth embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 6 shows the steps of a method according to a sixth embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 7 shows the steps of a method according to a seventh embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification; Figure 8 shows the steps of a method according to an eighth embodiment of the invention of configuring an optical communication network comprising an optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification;

Figure 9 is a schematic representation of an optical network element according to a ninth embodiment of the invention;

Figure 10 is a schematic representation of an optical network element according to a thirteenth embodiment of the invention; and

Figure 11 is a schematic representation of an optical communication network according to a fifteenth embodiment of the invention.

Detailed description

A first embodiment of the invention provides a method 10 of configuring an optical communication network. The network comprises an optical link comprising an optical fibre configured as a Raman amplifier. The Raman amplifier is configured for co- propagating Raman amplification.

The method 10 comprises:

a. selecting an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio, OSNR, penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier 12;

b. selecting an optical signal power, P_c , for at least one optical

communication signal to be transmitted across the optical link, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain 14;

c. generating a pump signal power control signal configured to cause an

optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power 16; and d. generating a signal power control signal configured to cause the optical communication signal to be delivered into the optical link at the selected optical signal power 18.

It will be appreciated by the person skilled in the art that an optical signal to noise ratio, OSNR, penalty caused by double Rayleigh scattering will always arise in a Raman amplifier but that below a certain gain threshold the OSNR penalty is negligible. That is to say, the OSNR penalty will not result in traffic carried by the optical communication signal being lost or incorrectly identified at optical receiver apparatus in the optical

communication network arranged to receive the optical communication signal.

For backward Raman amplification the effective noise figure gets better up to a certain gain, but after that it increases again because of the double Rayleigh scattering. So the gain threshold is set below the gain value at which the noise figure increase cancels all or most of the benefit of the Raman gain. For forward Raman amplification the noise figure is expected to increase with gain, but when double Rayleigh scattering starts to occur the noise figure increases further as a result. Q factor is the bit error rate, BER, expressed through the inverse of the erfc() function.

The steps of a method 20 of configuring an optical communication network according to a second embodiment of the invention are shown in Figure 2. The steps of the method 20 of this embodiment are similar to the steps of the method 10 of the first embodiment, with the following modifications. The same reference numbers are retained for corresponding steps.

In this embodiment, step b. comprises:

i. obtaining an optimal optical signal power, P_ni, for the optical communication signal at which the Q-factor of the optical communication signal is maximised in the absence of Raman amplification 22;

ii. calculating a nonlinear enhancement factor, NEF, being an optical signal power reduction required to the optimal optical signal power for said maximised saturated gain to avoid an increase in a nonlinear degradation of the optical signal when Raman amplification is applied to the optical communication signal 24; and

iii. calculating 26 an optimal value for the optical signal power, P_c , as P_ck = P_nl - NEF

A signal power control signal configured to cause the optical communication signal to be delivered into the optical link at the calculated optical signal power is then generated 28.

The steps of a method 30 of configuring an optical communication network according to a third embodiment of the invention are shown in Figure 3. The steps of the method 30 of this embodiment are similar to the steps of the method 20 of the second embodiment, with the following modifications. The same reference numbers are retained for corresponding steps.

In this embodiment, the optimal optical signal power in the absence of Raman amplification, P_ni, is obtained by determining a number of optical communication signals, N_ch, to be transmitted across the optical link. A pre-stored value of P_ni is then retrieved in dependence on the number of optical communication signals 32. It will be appreciated that the pre-stored value of P_ni may alternatively be retrieved in dependence on the type of the optical fibre or the modulation format of the optical communication signals, or a combination of any of these three parameters.

The steps of a method 40 of configuring an optical communication network according to a fourth embodiment of the invention are shown in Figure 4. The steps of the method 40 of this embodiment are similar to the steps of the method 20 of the second embodiment, with the following modifications. The same reference numbers are retained for corresponding steps.

In this embodiment, step a. comprises selecting an initial value for the optical signal power 42 for at least one optical communication signal to be transmitted across the optical link. The initial value of the optical signal power is manually set by an engineer in the field to maximise a Q-factor of the optical signal, for an initial value of the optical pump signal, also manually set by the engineer.

Step b. further comprises obtaining a value for the saturated gain, G, of the Raman amplifier at the manually set optical pump signal power and optical signal power. Step b. comprises calculating a nonlinear enhancement factor, NEF, for the manually set optical signal power 44. An optimal optical signal power in the absence of Raman amplification, P„i, is then calculated 46 as P_nl = P_ch + NEF The steps of a method 50 of configuring an optical communication network according to a fifth embodiment of the invention are shown in Figure 5. The steps of the method 50 of this embodiment are similar to the steps of the method 30 of the third embodiment, with the following modifications. The same reference numbers are retained for corresponding steps.

In this embodiment, a saturated gain of the Raman amplifier is measured and the nonlinear enhancement factor is calculated 52 as NEF = M * G . is a scaling factor in the range 0.7-0.8. The value of M may be calibrated, by measurement or modelling, for a specific Raman pumping scheme which is to be used.

The steps of a method 60 of configuring an optical communication network according to a sixth embodiment of the invention are shown in Figure 6. The steps of the method 60 of this embodiment are similar to the steps of the method 30 of the third embodiment, with the following modifications. The same reference numbers are retained for corresponding steps.

In this embodiment, the saturated gain, G(z), is calculated 62 as

G₀ (z)(P_p + S)

G(z) , where z is a distance along the optical link, S is the normalised P_p + G₀ (z)S

The nonlinear enhancement factor is calculated 64 as the ratio of the effective len ths of the Raman amplifier with and without Raman amplification, which can be given

The steps of a method 70 of configuring an optical communication network according to a seventh embodiment of the invention are shown in Figure 7. The configuration of an optical communication network with co-propagating Raman amplification requires setting of the optical pump signal power and the optical signal power for the optical communication signals to be transmitted across the network. This embodiment enables automatic selection of the optical signal power.

The method 70 comprises ramping the optical pump signal power up to the maximum power, P_max, that the respective optical signal pump sources are able to generate 72. The optical signal power is set to a minimum operating value, P_ch _min. A number of system parameters of the optical communication network are then acquired 74: the number of optical communications signals, N_ch is determined; the currently set optical signal power, Pc is determined; the currently set optical pump signal power, P_p, is determined; and the optimal optical signal power in the absence of Raman amplification, P_ni, is retrieved in dependence on N_ch.

The method 70 then proceeds in two different ways depending upon whether the saturated gain, G, of the Raman amplifier is measured 76.

If the gain is measured, the nonlinear enhancement factor, NEF, is calculated 78 as NEF = M * G , where G is the measured saturated gain and M is typically approximately 0.8. This equation for NEF has been empirically obtained and proved by simulation. An optimal value for the optical signal power is then calculated 78 as P_ch = P_nl - NEF .

If the gain is not measured it is determined from other parameters of the optical communication network, using a gain saturation model of the Raman amplifier, for single order Raman pumping, which takes into account both gain saturation of the optical pump signal and nonlinear impairments as follows.

G₀ (z)(P_p + S)

The saturated gain, G(z), is calculated 86 as G(z) = , where z is a

P_p + G₀ (z)S

distance along the optical link, S is the normalised total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier. This equation enables the saturated gain to be calculated while taking into account gain saturation by the optical communication signals.

The unsaturated gain of the Raman amplifier is given by

G₀ (z) = exp((i^, + S)(l— e ^az )) , where a is the loss coefficient of the optical link at a wavelength of the optical communication signal.

/ 8 P (°)

the optical communication signal frequency,^ is the optical pump signal frequency, g_¾ is the Raman gain of the fibre, P_p(0) is the normalised optical pump signal power at z=0, a_p is the loss coefficient of the fibre at the pump signal frequency, and A_e/f is the effective area of the fibre. The normalised total optical signal power is given by S = S ^R P ^s (0) where

P_s(0)is the optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency.

The enhancement of nonlinear impairments due to distributed Raman gain is taken into account by the nonlinear enhancement factor, NEF, which is defined as the ratio of the Effective Lengths of the optical link with and without Raman amplification:

The nonlinear enhancement factor says how much to lower the optical signal power by in order to not incur an increase in nonlinear impairments. The nonlinear enhancement factor is calculated 86 in this embodiment as NEF

Given the optimal optical signal power in the absence of Raman amplification, P_ni, an optimal value for the optical signal power in the presence of co-propagating Raman amplification is then calculated 86 as P_ch = P_nl - NEF .

The Raman link budget improvement is given by P_ch+G-P_ni = G-NEF.

Following selection of the optical signal power, P_c , the method 70 proceeds by checking 80 whether the value of the saturated gain, G, determined in the previous steps by measurement or calculation is greater than the gain threshold, G_max, of the Raman amplifier plus a hysteresis parameter, Δ. The hysteresis parameter, Δ, is selected to be larger than the network tolerance to avoid oscillations in the optical signal power but is small enough so that G is not selected to have a value which would degrade the performance, i.e. reduce the Q-factor, of the communication network. Δ typically has a value of approximately 0.5dB in this embodiment. If G > G_max + Δ, the optical pump signal power is reduced 88, in order to reduce the saturated gain, and the method repeats the loop of determining G, NEF and P_ch.

If G≤ G_max + Δ, the method 70 proceeds to checking 82 whether G < G_max - Δ and whether P_p < P_max, in case the optical pump signal power has previously been reduced and can now be increased again. If both conditions are met, the optical pump signal power is increased 90. If either condition is not met, the optical pump signal power is maintained and the optical signal power is set 84 to the value, P_ch, in the previous steps 78, 86. G_max may be a pre-selected value which is set by a manufacturer of an optical pump signal source configured to generate the optical pump signal. Alternatively, G_max may be determined by characterisation measurements of the optical pump signal source configured to generate the optical pump signal.

Where the saturated gain is measured, the method 70 will typically require the steps of the pump signal power being reduced 88 or increased 90 and the optimal value for the optical signal power being recalculated 78 to be carried out at least once.

If the number of optical communication signals, N_ch, changes the method 70 may be recommenced at the step of acquiring system parameters 74.

With this method the optical signal power launched into the optical link may be optimised, accounting for both pump depletion and nonlinear impairments. Furthermore, no Raman pump signal power is wasted since the method of this embodiment sets the optical pump signal power to a maximum available value. Provided that the gain, G, is not higher than the gain threshold, G_max, to avoid secondary effects including Double Rayleigh Scattering, the method of this embodiment enables best performance for an optical communication network to achieved by configuring the Raman amplifier to have the maximum acceptable gain, by setting the pump sources to their maximum pump signal power, and setting the optical signal power such that nonlinear degradation is not caused during propagation of the optical communication signals across the optical link.

In fact, it has been found that for a given nonlinear penalty, increasing the Raman amplifier gain requires an optical signal power reduction of less than 1 dB per dB of gain increase, thus improving the optical signal to noise ratio, ONSR, and reach of the optical communication network. The higher the gain, the higher the benefit.

It will be appreciated by the person skilled in the art that an OSNR penalty caused by double Rayleigh scattering, DRS, will always arise in a Raman amplifier, and an OSNR penalty due to relative intensity noise transfer, RIN, may also arise. The gain threshold, G_max, is selected such that the OSNR penalty due to DRS and RIN is negligible, which here means less than 0.5dB due to DRS and less than 0.1 dB due to RIN.

For backward Raman amplification the effective noise figure gets better up to a certain gain, but after that it increases again because of the DRS. So the gain threshold is set below the gain value at which the noise figure increase cancels all or most of the benefit of the Raman gain. For forward Raman amplification the noise figure is expected to increase with gain, but when DRS starts to occur the noise figure increases further as a result. Where the optical communication signal has a high power before Raman amplification the main source of OSNR is expected to be RIN transfer.

The steps of a method 100 of configuring an optical communication network according to an eighth embodiment of the invention are shown in Figure 8. The method 100 of this embodiment is similar to the method 70 of Figure 7, with the following modifications. The same reference numbers are retained for corresponding features.

In this embodiment, the method 100 commences with an engineer manually setting initial values 102 for the optical signal power, P_c , and the optical pump signal power, P_p.

The optimal optical signal power in the absence of Raman amplification, P_ni, is then calculated 104 by determining the saturated gain, G(z) and nonlinear enhancement factor, NEF, for the initially set values for P_c and P_p, using the following equations, as described above:

G₀ (z)(P_B + S)

G(z) = ° ^p

P_p + G₀ (z)S

G₀ (z) = _Qxp((P_p + S)(l - e-^az ))

P_ni is calculated as P_nl = P_ch + NEF

An optical communication network element, ONE, 110 according to a ninth embodiment of the invention is shown in Figure 9. The ONE 110 comprises an optical pump signal source 112, optical communication signal power control apparatus 114, an output 116 and a controller 122.

The output 116 is configured to output at least one optical communication signal 118 and a co-propagating optical pump signal 120 to be delivered to an optical link comprising an optical fibre configured as a Raman amplifier.

The controller 122 is configured to:

a. select an optical pump signal power, P_p, such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier;

b. select an optical signal power, P_c , for the at least one optical communication signal, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain;

c. generate a pump power control signal 124 configured to cause the optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power; and

d. generate a signal power control signal 126 configured to cause the optical communication signal power control apparatus to set the optical signal power of the at least one optical communication signal to the selected optical signal power.

It will be appreciated by the person skilled in the art that double Rayleigh scattering, DRS, will always be caused in a Raman amplifier but that below a certain threshold the optical signal to noise ratio, OSNR, penalty caused by the DRS is negligible. That is to say, the OSNR penalty will not result in traffic carried by the optical

communication signal being lost or incorrectly identified at optical receiver apparatus in the optical communication network arranged to receive the optical communication signal.

A tenth embodiment of the invention provides an optical communication network element, ONE, which is similar to the ONE 110 shown in Figure 9 and which will be described with reference to Figure 9.

In this embodiment, the controller 122 is configured to, in step b.:

i. obtain an optimal optical signal power, P_ni, for the optical communication signal at which the Q-factor of the optical communication signal is maximised in the absence of Raman amplification;

P_ck = P_nl - NEF An eleventh embodiment of the invention provides an optical communication network element, ONE, which is similar to the ONE 110 shown in Figure 9 and which will be described with reference to Figure 9.

In this embodiment, the controller 122 is configured in step b.i. to determine a number of optical communication signals, N_ch, to be delivered to the optical link. The controller is configured to obtain the optimal optical signal power, P_ni, by retrieving a pre- stored value of P_ni in dependence on the number of optical communication signals. It will be appreciated that the controller 122 may alternatively be configured to retrieve a pre- stored value of P_ni in dependence on the type of the optical fibre or the modulation format of the optical communication signals, or a combination of any of these three parameters.

A twelfth embodiment of the invention provides an optical communication network element, ONE, which is similar to the ONE 110 shown in Figure 9 and which will be described with reference to Figure 9.

In this embodiment, the controller 122 is configured to operate following an engineer manually setting initial values for the optical signal power, P_c , and the optical pump signal power, P_p.

The controller is configured, in step b., to obtain a value for the saturated gain, G, of the Raman amplifier at the selected initial value for the optical pump signal power and the selected initial value for the optical signal power. The controller is configured to calculate a nonlinear enhancement factor, NEF, for the selected initial value for optical signal power and to calculate an optimal optical signal power in the absence of Raman amplification, P_ni, as

P_nl = P_ck + NEF

A thirteenth embodiment of the invention provides an optical communication network element, ONE, 130 as shown in Figure 10. The ONE 130 of this embodiment is similar to the ONE 110 shown in Figure 9, with the following modifications. The same reference numbers are retained for corresponding features.

In this embodiment, the ONE 130 comprises gain measurement apparatus 132 configured to measure the saturated gain of the Raman amplifier. The controller 122 is configured to receive the measured saturated gain, G, and to calculate the nonlinear enhancement factor as NEF = M * G . is in the range 0.7-0.8. A fourteenth embodiment of the invention provides an optical communication network element, ONE, which is similar to the ONE 1 10 shown in Figure 9 and which will be described with reference to Figure 9.

In this embodiment, the controller 122 is configured to calculate saturated gain,

G₀ (z)(P_p + S)

G(z), as G(z) = , where z is a distance along the optical link, S is the

P_p + G₀ (z)S

normalised total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier. The unsaturated gain is given by

/ 8 P (°)

the optical communication signal frequency,^ is the optical pump signal frequency, g_¾ is the Raman gain of the fibre, P_p(0) is the normalised pump signal power at z=0, a_p is the loss coefficient of the fibre at the pump signal frequency, and A_e/f is the effective area of

8 P (0)

the fibre. The normalised total optical signal power is given by S = ^{R s} where P_s(0) is the optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency.

The controller is configured to calculate the nonlinear enhancement factor as the ratio of the effective lengths of the Raman amplifier with and without Raman

amplification:

In this embodiment, the controller 122 is configured to calculate the nonlinear enhancement factor as NEF

A fifteenth embodiment of the invention provides an optical communication network 140 as shown in Figure 1 1. The optical communication network comprises an optical link 142 and an optical communication network element, ONE, 1 10. The optical link 142 comprises an optical fibre configured as a Raman amplifier. The Raman amplifier is configured for co -propagating Raman amplification. The optical communication network element 110 is as shown in Figure 9, according to any one of the ninth to fourteenth embodiments.

It will be appreciated by the person skilled in the art that an OSNR penalty caused by double Rayleigh scattering will always be caused in a Raman amplifier but that below a certain threshold it is negligible. That is to say, the OSNR penalty will not result in traffic carried by the optical communication signal being lost or incorrectly identified at optical receiver apparatus in the optical communication network arranged to receive the optical communication signal.

A sixteenth embodiment of the invention provides a data carrier having computer readable instructions embodied therein. The computer readable instructions are for providing access to resources available on a processor and comprise instructions to cause the processor to perform any of the steps of the method of configuring an optical communication network comprising an optical link comprising a Raman amplifier configured for co-propagating Raman amplification as described above in relation to Figures 1 to 8.

Claims

1. A method of configuring an optical communication network comprising an

optical link comprising an optical fibre configured as a Raman amplifier configured for co-propagating Raman amplification, the method comprising: a. selecting an optical pump signal power, P_p, of the Raman amplifier such that a saturated gain, G, of the Raman amplifier is at a maximum operating level without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier;

b. selecting an optical signal power, P_c , for at least one optical

communication signal to be transmitted across the optical link, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain;

c. generating a pump signal power control signal configured to cause an

optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power; and

d. generating a signal power control signal configured to cause the optical communication signal to be delivered into the optical link at the selected optical signal power.

2. A method as claimed in claim 1, wherein b. comprises:

i. obtaining an optimal optical signal power, P_ni, for the optical

ii. calculating a nonlinear enhancement factor, NEF, being an optical signal power reduction required to the optimal optical signal power for said maximised saturated gain to avoid an increase in a nonlinear degradation of the optical signal when Raman amplification is applied to the optical communication signal; and iii. calculating an optimal value for the optical signal power, P_c , as ch P n.l NEF

3. A method as claimed in claim 2, wherein the optimal optical signal power in the absence of Raman amplification, P_ni, is obtained by determining a number of optical communication signals, N_ch, to be transmitted across the optical link and retrieving a pre-stored value of P_ni in dependence on at least one of the number of optical communication signals, a characteristic of the optical fibre and a modulation format of the optical communication signals.

4. A method as claimed in claim 2, wherein a. comprises selecting an initial value for the optical signal power and wherein b. further comprises obtaining a value for the saturated gain, G, of the Raman amplifier at the selected optical pump signal power and the selected optical signal power, and calculating a nonlinear enhancement factor, NEF, for the selected optical signal power, and calculating an optimal optical signal power in the absence of Raman amplification, P_ni, as

P n.l P_ch + NEF

A method as claimed in any of claims 2 to 4, wherein the saturated gain is obtained by measurement and the nonlinear enhancement factor is calculated as NEF = M * G , where Mis in the range 0.7-0.8.

A method as claimed in any of claims 2 to 4, wherein the saturated gain, G(z), is

G₀ (z)(P_p + S)

calculated as G(z) , where z is a distance along the optical

P_p + G₀ (z)S

link, S is the normalised total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier.

7. A method as claimed in claim 6, wherein Go(z) is given by

G₀ (z) = Qxp((P_p + S)(l - e^~az )) , where a is the loss coefficient of the optical link at a wavelength of the optical communication signal, and wherein the nonlinear enhancement factor is calculated as the ratio of the effective lengths of the Raman amplifier with and without Raman amplification and f_sg_RP_P (0) wherein the normalised optical pump signal power is given by P

fp^ap^Aeff where f_s is the optical communication signal frequency,^, is the optical pump signal frequency, g¾ is the Raman gain of the fibre, P_p(0) is the normalised pump signal power at z=0, a_p is the loss coefficient of the fibre at the pump signal frequency, and A_e/f is the effective area of the fibre, and the normalised

S P (0)

total optical signal power is given by S = ^{R s} where P_s(0) is the optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency.

A method as claimed in claim 6 or 7 wherein the nonlinear enhancement factor is calculated as NEF =

9. An optical communication network element comprising:

an optical pump signal source;

optical communication signal power control apparatus;

an output configured to output at least one optical communication signal and a co-propagating optical pump signal to be delivered to an optical link comprising an optical fibre configured as a Raman amplifier; and

a controller configured to:

communication signal, the optical signal power being selected to maximise a Q-factor of the optical signal when the Raman amplifier is configured for said maximum operating level of the saturated gain; c. generate a pump power control signal configured to cause the optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power; and

d. generate a signal power control signal configured to cause the optical

10. An optical communication network element as claimed in claim 9, wherein the controller is configured to, in step b.:

i. obtain an optimal optical signal power, P_ni, for the optical

P_ck = P_nl - NEF

11. An optical communication network element as claimed in claim 10, wherein the controller is configured in step b.i. to determine a number of optical communication signals, N_ch, to be delivered to the optical link and is configured to obtain the optimal optical signal power, P_ni, by retrieving a pre-stored value of P_ni in dependence on at least one of the number of optical communication signals, a characteristic of the optical fibre and a modulation format of the optical communication signals.

12. An optical communication network element as claimed in claim 10, wherein the controller is configured in step b. to obtain a value for the saturated gain, G, of the Raman amplifier at the selected optical pump signal power and the selected optical signal power, and is configured to calculate a nonlinear enhancement factor, NEF, for the selected optical signal power and to calculate an optimal optical signal power in the absence of Raman amplification, P_ni, as

P_nl = P_ck + NEF

13. An optical communication network element as claimed in any of claims 10 to

12, wherein the communication network element comprises gain measurement apparatus configured to measure the saturated gain of the Raman amplifier and the controller is configured to receive the measured saturated gain and to calculate the nonlinear enhancement factor as NEF = M * G , where M is in the range 0.7-0.8.

14. An optical communication network element as claimed in any of claims 10 to

12, wherein the controller is configured to calculate saturated gain, G(z), as G₀ (z)(P_p + S)

G(z) = , where z is a distance along the optical link, S is the

P_p + G₀ (z)S

normalised total optical signal power, P_p is the normalised optical pump signal power, and Go(z) is the unsaturated gain of the Raman amplifier.

15. An optical communication network element as claimed in claim 14, wherein

Go(z) is given by G₀ (z) = Qxp((P_p + S)(l - e^~az )) , where a is the loss coefficient of the optical link at a wavelength of the optical communication signal, and wherein the controller is configured to calculate the nonlinear enhancement factor as the ratio of the effective lengths of the Raman amplifier with and without Raman amplification and wherein the normalised optical

/ 8 P (°)

pump signal power is given by P = where f_s is the optical communication signal frequency,^, is the optical pump signal frequency, g_¾ is the Raman gain of the fibre, P_p(0) is the normalised pump signal power at z=0, OLp is the loss coefficient of the fibre at the pump signal frequency, and A_e/fis the effective area of the fibre, and the normalised total optical signal power is given by S = S ^R P ^s (0) where P_s(0) is the optical signal power at z=0, and a_s is the loss coefficient of the fibre at the optical communication signal frequency.

An optical communication network element as claimed in claim 14 or 15, wherein the controller is configured to calculate the nonlinear enhancement factor as NEF

An optical communication network comprising:

an optical link comprising an optical fibre configured as a Raman amplifier configured for co -propagating Raman amplification; and

an optical communication network element comprising:

an optical pump signal source;

optical communication signal power control apparatus;

an output configured to output at least one optical communication signal and a co-propagating optical pump signal to be delivered to an optical link

comprising a Raman amplifier; and

a controller configured to:

a. select an optical pump signal power, P_p, such that a saturated gain, G, of the Raman amplifier is maximised without exceeding a gain threshold above which an optical signal to noise ratio penalty caused by double Rayleigh scattering is no longer negligible in the Raman amplifier;

b. select an optical signal power, P_c , for the at least one optical

communication signal, the optical signal power being selected to maximise a Q-factor of the optical signal for said maximised saturated gain of the Raman amplifier; c. generate a pump power control signal configured to cause the optical pump signal source of the Raman amplifier to generate an optical pump signal having the selected optical pump signal power; and

d. generate a signal power control signal configured to cause the optical

A data carrier having computer readable instructions embodied therein, the computer readable instructions being for providing access to resources available on a processor and the computer readable instructions comprising instructions to cause the processor to perform any of the steps of the method as claimed in any of claims 1 to 8 of configuring an optical communication network comprising an optical link comprising a Raman amplifier configured for co- propagating Raman amplification.